Metraipolis › archive › dc_95db756bc692

Training-Free Active Test-Time Adaptation for Vision-Language Models — arXiv:2611.09084

Source: arXiv (en)
Original: https://arxiv.org/abs/2611.09084
License (as archived): CC0-1.0
Archived: 2026-11-09T22:34:49.563Z
SHA-256 of the archived text: afec0b8810af853101fc5d4c43f8ce2b1ab8e945773ae6a3da0bb1a90f3a9e61
Stable URL of this page: https://metraipolis.com/documento/dc_95db756bc692

How to cite

Training-Free Active Test-Time Adaptation for Vision-Language Models — arXiv:2611.09084 — arXiv (en), https://arxiv.org/abs/2611.09084 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_95db756bc692 (Metraipolis archive dc_95db756bc692, sha256 afec0b8810af853101fc5d4c43f8ce2b1ab8e945773ae6a3da0bb1a90f3a9e61)

BibTeX

@misc{dc95db756bc692,
  title = {Training-Free Active Test-Time Adaptation for Vision-Language Models — arXiv:2611.09084},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.09084},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_95db756bc692; sha256 afec0b8810af853101fc5d4c43f8ce2b1ab8e945773ae6a3da0bb1a90f3a9e61},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09084. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Test-time adaptation (TTA) for vision-language models (VLMs) is usually studied in a fully unsupervised regime, where adaptation relies on pseudo-labels, confidence scores, or entropy computed by the model itself. This is a severe restriction on hard shifted samples: the model is asked to repair its own uncertain predictions without any external evidence. At the other extreme, supervised online adaptation assumes labels can directly train or correct the model. We study the intermediate deployment regime between these two extremes. In Active Test-time Adaptation for VLMs (Active VLM-TTA), a pretrained VLM may spend an explicit labeling budget on selected test samples, but the prediction for the current sample must be emitted before its label can be used; queried labels are therefore costed, delayed evidence for future samples rather than free correction for the present one. Under this future-only constraint, the central question is how to reuse sparse labels conservatively. We instantiate this setting with ACTive VLM-TTA for Online Retrieval (ACTOR), a training-free plug-in module that routes uncertain samples among base prediction, label query, and memory-based correction using a consistency-and-reliability check over previously queried labels. Across ten cross-domain benchmarks and a domain-transfer benchmark based on ImageNet, ACTOR consistently improves strong VLM-TTA baselines on ResNet and ViT backbones with negligible inference overhead. Our implementation is publicly available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.09084 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-09T22:30:47.318Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).